Serving the Quantitative Finance Community

 
frolloos
Topic Author
Posts: 752
Joined: September 27th, 2007, 5:29 pm
Location: Netherlands

Gershon / superderivatives vol model

May 11th, 2017, 2:43 pm

Is it true that the Superderivatives pricing model (developed by David Gershon), which supposedly matches IDB market prices (also for exotics), has been used for quite some time now by sell-side?

I cam across an article which suggests the above is the case.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Re: Gershon / superderivatives vol model

May 12th, 2017, 11:43 am

For what products?  Last I looked, they had some glaring issues and, emphatically, didn't match IDB mkt prices for some rates products.  To be fair, this was some time ago.
 
frolloos
Topic Author
Posts: 752
Joined: September 27th, 2007, 5:29 pm
Location: Netherlands

Re: Gershon / superderivatives vol model

May 12th, 2017, 2:18 pm

The article / paper mentions swaptions, FX DNTs and plain vanillas. That is quite comprehensive. And if true that Gershon / SD model can match IDB for said instruments, then I suppose we can all pack-up and go home :-D Which I doubt, but possible of course.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Re: Gershon / superderivatives vol model

May 12th, 2017, 2:21 pm

When I looked it was OK for swaptions, mostly...  However, it had pretty monumental issues with caps and floors.
 
User avatar
VolMaster
Posts: 20
Joined: December 5th, 2009, 8:48 am

Re: Gershon / superderivatives vol model

June 12th, 2017, 4:30 am

As someone who worked at SuperDerivatives for few years, I can comfortably tell you that their FX model is simply Vanna-Volga model. It's commonly used for G10 FX, while in recent years more and more sell-side and option desks are leaning toward stochastic-local vol or SABR model...

Just my 2cents
 
frolloos
Topic Author
Posts: 752
Joined: September 27th, 2007, 5:29 pm
Location: Netherlands

Re: Gershon / superderivatives vol model

June 12th, 2017, 3:11 pm

Thanks - yes, that was the idea I got as well, that it is a variation on "cost of Greek" method, of which there are a few out there.